sign in sign up
<<
Home , Fibrinolysis, Plasminogen activator, Thrombolysis, Urokinase

JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY VOL.68,NO.19,2016

ª 2016 PUBLISHED BY ELSEVIER ON BEHALF OF THE ISSN 0735-1097/$36.00

AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION http://dx.doi.org/10.1016/j.jacc.2016.07.780

THE PRESENT AND FUTURE

REVIEW TOPIC OF THE WEEK Therapeutic How Efficacy and Safety Can Be Improved

Victor Gurewich, MD

ABSTRACT

Therapeutic fibrinolysis has been dominated by the experience with tissue-type (t-PA), which proved little better than in acute . In contrast, endogenous fibrinolysis, using one-thousandth of the t-PA concentration, is regularly lysing fibrin and induced In Myocardial Infarction flow grade 3 patency in 15% of patients with acute myocardial infarction. This efficacy is due to the effects of t-PA and plasminogen activator (uPA). They are complementary in fibrinolysis so that in combination, their effect is synergistic. Lysis of intact fibrin is initiated by t-PA, and uPA activates the remaining plasminogens. Knockout of the uPA gene, but not the t-PA gene, inhibited fibrinolysis. In the clinic, a minibolus of t-PA followed by an infusion of uPA was administered to 101 patients with acute myocardial infarction; superior infarct artery patency, no reocclusions, and 1% mortality resulted. Endogenous fibrinolysis may provide a paradigm that is relevant for therapeutic fibrinolysis. (J Am Coll Cardiol 2016;68:2099–106) © 2016 Published by Elsevier on behalf of the American College of Cardiology Foundation.

n occlusive intravascular triggers fibrinolysis, as shown by it frequently not being A the cardiovascular diseases that are the lead- identified specifically in publications on clinical ing causes of death and disability worldwide. fibrinolysis. The only pharmacological means to remove the The anticipated benefits of t-PA were never ful- thrombus is fibrinolysis. There are 2 plasminogen filled, and it has been largely replaced by percuta- activators in the responsible for fibrinolysis: neous coronary intervention (PCI) as the treatment of tissue-type plasminogen activator (t-PA) and uroki- choice for AMI in urban centers in the United States, nase plasminogen activator (uPA). They exist in although it is still used in rural areas too far from a both single-chain and -cleaved 2-chain forms. catheterization facility. As pre-treatment, or facili- With t-PA, both forms have the same activity (1), tated PCI, t-PA significantly increased both throm- whereas the single-chain form of uPA is a proenzyme botic and hemorrhagic complications of PCI (7),and (pro-uPA) and the 2-chain form is an active itsusehasbeendiscouraged.Inischemicstroke,t-PA (2-chain uPA, the enzymatic form of uPA [tc-uPA]) (2). is used in only about 5% of patients (8), and in venous The fibrin specificity of t-PA is mediated by high thromboembolism, a hemorrhagic risk has limited its fibrin affinity (3,4), which promotes plasminogen use to patients with hemodynamic instability due to activating activity about a thousand-fold (5).Itwas the emboli (9). the first fibrin-specific plasminogen activator, and The enzymatic form of uPA, tc-uPA, has been became the first recombinant on the market known since the 1940s as a nonspecific activator, like (6). The clinical experience with t-PA in acute streptokinase (SK), which caused bleeding complica- Listen to this manuscript’s myocardial infarction (AMI), ischemic , and tions. The single-chain uPA proenzyme, pro-uPA, was audio summary by venous thromboembolism has been extensive. As a not identifiedandisolateduntil1980(10).Ithasno JACC Editor-in-Chief result, t-PA became essentially equated with fibrin affinity, but is nevertheless fibrin specific (11). Dr. Valentin Fuster.

From the Vascular Research Laboratory, Mount Auburn Hospital, Cambridge, Massachusetts, and Harvard Medical School, Cam- bridge, Massachusetts. Dr. Gurewich has reported that he has no relationships relevant to the contents of this paper to disclose.

Manuscript received July 14, 2016; accepted July 20, 2016. 2100 Gurewich JACC VOL. 68, NO. 19, 2016 Improving Fibrinolysis NOVEMBER 8, 2016:2099– 106

ABBREVIATIONS Being a proenzyme, it remains inert in blood A special challenge for primary PCI treatment is AND ACRONYMS until it reaches a clot, a pharmacologically reducing the time to reperfusion. Reduction in AMI advantageous property. Unfortunately, this mortality is greatest if reperfusion is accomplished AMI = acute myocardial infarction property is undermined by dose-dependent within 1 to 2 h of symptom onset (27),andin plasma instability, because it activates patients who received reperfusion treatment within ICH = intracranial hemorrhage plasma plasminogen at certain therapeutic 70 min of symptom onset, the mortality was 1.2% PCI = percutaneous coronary intervention concentrations, and plasmin converts pro- (28). In animal models, the longer the coronary uPA to tc-uPA. For example, in a study occlusion, the less salvageable myocardium remains pro-uPA = proenzyme urokinase plasminogen where pro-uPA and SK were compared in (29). Therefore, pre-treatment was tested. activator AMI, bleeding complications occurred in 14% FACILITATED PCI WITH t-PA SK = streptokinase of the pro-uPA patients, of whom 4% tc-uPA = 2-chain urokinase required transfusions and 2 patients had an Although t-PA pre-treatment doubled the number of plasminogen activator, the intracranial hemorrhage (ICH) (12).Atphysi- enzymatic form of uPA patients with initial TIMI flow grade 3 patency, the ological and lower therapeutic concentra- TIMI = Thrombolysis In final TIMI flow grade 3 patency rates did not differ in tions, pro-uPA is stable in plasma and is an Myocardial Infarction the 2 groups (89% vs. 88%), consistent with t-PA’s essential component of the endogenous t-PA = tissue-type significant rethrombosis rate (17). In addition, more plasminogen activator fibrinolytic system. bleeding, reinfarctions, and occurred with uPA = urokinase plasminogen FIBRINOLYSIS WITH t-PA facilitated PCI (30). By contrast, with pro-uPA, neither activator a prothrombotic effect nor reocclusions were found in The U.S. Food and Administration approved AMI (31), suggesting that the complications were t-PA for AMI treatment in 1987 but, to show that it related to t-PA rather than to fibrinolysis. reduced mortality compared with SK, additional It was concluded that t-PA facilitated PCI offered clinical trials were performed starting in 1990. Three no benefit, an outcome lamented in a commentary in megatrials with an unprecedented total of 95,740 a prominent journal as “Facilitated angioplasty: patients were required before a significant p value paradise lost” (32), because there is no alternative to was reached for a mortality difference (13–15).This shorten the time to reperfusion. was in only 1 of 4 groups in the last trial, in which an Interestingly, these studies found that in 15% of accelerated rate of t-PA infusion induced a significant patients with ST-segment elevation myocardial 30-day mortality difference (6.3% vs. 7.0%) (15).Also infarction, TIMI flow grade 3 was already found at the unexpected was the finding that the ICH rate was time of the primary PCI (33), and these patients fared significantly higher with t-PA than SK, despite the significantly better (24,25). This illustrates the effi- nonspecific and indirect mode of action of SK. These cacy of endogenous fibrinolysis in which the t-PA results did not stand up to a Bayesian analysis, which concentration is $1,000-fold lower than the thera- concluded, “the clinical superiority of t-PA over SK peutic one. remains uncertain” (16). t-PA FIBRINOLYSIS INISCHEMICSTROKE As heralded by these trials, t-PA’s limited efficacy— a16 10% reocclusion rate (17)—prothrombotic he- In ischemic stroke, a t-PA dose reduction was matologic effects, such as generation and required due to a 20% symptomatic ICH incidence activation (18–22), and bleeding side effects, whentheusualAMIdosewasused(34).Asaresult, such as ICH, led to alternative therapies. the t-PA recanalization decreased 30% to 40%, PCI VERSUS t-PA making it ineffective for larger thrombi in proximal occlusions (35,36).Evenatthereduceddose,the In a review of 23 PCI trials involving 7,739 patients, symptomatic ICH rate was 6% to 7% (37). 3,867 were randomized to treatment with fibrinolysis In the largest trial, 1,500 patients per group were (2,939 with t-PA, 928 with SK) and 3,872 patients to randomized to t-PA or standard care and treated within primary PCI (23).Asignificantly better outcome was 6 h. The proportion of patients alive and living inde- obtained with PCI for both the short-term and long- pendently at 6 months was 37% with t-PA and 35% term outcomes. The post-treatment Thrombolysis In for the controls. Although insignificant, a secondary Myocardial Infarction (TIMI) flow grade 3, an impor- ordinal analysis found a significant (p < 0.001) favor- tant predictor of good outcome (24,25),wasalso able shift in the functional score in the t-PA–treated significantly better with PCI than in the GUSTO (Global patients, but more patients died with t-PA in the first Utilization of Streptokinase and Tissue Plasminogen 7days(11%vs.7%;p¼ 0.001) (38).Theauthors Activator for Occluded Coronary Arteries) trial (26). concluded that early treatment with t-PA should be JACC VOL. 68, NO. 19, 2016 Gurewich 2101 NOVEMBER 8, 2016:2099– 106 Improving Fibrinolysis

encouraged, but others were more skeptical, including Catheter placement within the thrombus is a prereq- the American Academy of Emergency Medicine. A uisite for this indication, and a positive guidewire test 2002 position statement concluded that “evidence has been found to be a favorable prognostic sign for regarding the efficacy, safety, and applicability of tPA lysis (45).Successfullysis(>75% thrombus removal) for acute ischemic stroke is insufficient to warrant its was achieved in 72% of 119 patients; reocclusions classification as standard of care” (39).Althought-PAis occurred within 30 days in 13%, amputations were used in only about 5% of patients with ischemic stroke required in 6%, and life-threatening complications, overall (8), it is used in a higher percentage of patients including bleeding and stroke, occurred in 12% (46). in specialized centers. The potential benefitoft-PAfor Bleeding complications remain the main deterrent to reperfusion in stroke has been established, but efficacy more widespread use of fibrinolysis for these and safety need to be improved. indications. More recently, endovascular procedures are being used more frequently, with 3 trials published in the LESSONS FROM ENDOGENOUS FIBRINOLYSIS first 2 months of 2015 (40–42). Only patients with proximal thrombi in principal arteries in the anterior There is a noteworthy contrast between the limited circulation are currently eligible. therapeutic efficacy of t-PA and its efficacy in the endogenous fibrinolytic system, especially because , DEEP VEIN most of the t-PA in blood is in an inactive complex , AND LIMB with its inhibitor, plasminogen activator inhibitor-1 (47). Nevertheless, TIMI flow grade 3 patency in 15% PULMONARY EMBOLISM. Because of AMI patients (33) was induced, which compares provide effective and safe treatment for the majority with 45% of patients at 24 h by therapeutic t-PA at of patients, fibrinolysis is currently limited to hemo- a 1,000-fold higher concentration (26). In addition, dynamically unstable patients at risk of death (9).Ina endogenous fibrinolysis degrades fibrin regularly, study of 1,006 such patients randomized to antico- as seen by the fibrin degradation product, D-dimer agulants or fibrinolysis with mutant t-PA, fibrinolysis (a fibrinolysis-derived fibrin degradation product), reduced the risk of hemodynamic decompensation which has a normal plasma concentration of 110 to and death (2.6% vs. 5.6%), but increased the risk of 250 ng/ml. Only in the presence of an unusually major hemorrhage (6.3% vs.1.2%) and stroke (2.4% vs. potent autoantibody was the D-dimer 0.2%). It was concluded that, “great caution is war- level suppressed to 6 to 33 ng/ml (48), establishing ranted when considering current fibrinolytic therapy that D-dimer in plasma is derived from fibrinolysis of for this indication” (43). intravascular fibrin, which was suppressed by the . In the CaVenT (Catheter- anticoagulant. Directed Venous Thrombolysis in Acute Iliofemoral Therefore, the efficacy of endogenous fibrinolysis Vein Thrombosis) study of post-thrombotic is of interest due to its relevance as a model for morbidity, 209 patients with iliofemoral deep vein therapeutic fibrinolysis. The only difference between thrombosis were randomized to catheter-directed the 2 systems is that the endogenous system uses fi brinolysis with local infusion of t-PA (20 mg over t-PA in combination with uPA. Both activators are in “ ” 24 h) compared with alone. A relevant blood, and there is a body of evidence that both are 14.4% reduction of post-thrombotic syndrome was required for optimal fibrinolytic efficacy. Moreover, obtained, accompanied by a 21.7% bleeding risk at contrary to expectations, uPA, rather than t-PA, was puncture sites and into the calf, causing compartment identified as the dominant activator in fibrinolysis. fi syndrome (44). Therefore, a modest ef cacy was For example, 2 different gene knockout studies shown, accompanied by bleeding complications as in showed that eliminating t-PA had no apparent effect the experience described in the preceding text. The on lysis of an intravascular clot and did not induce largest study, ATTRACT (Acute : fibrin deposition, whereas knocking out uPA caused Thrombus Removal With Adjunctive Catheter- significant inhibition of clot lysis and some fibrin Directed Thrombolysis), is ongoing. deposition. Knocking out both genes had a signifi- ACUTE LIMB ISCHEMIA. Despite local delivery of the cantly stronger effect, suppressing lysis entirely and fibrinolytic treatment, the usual contraindications causing extensive fibrin deposition (49,50).There- due to bleeding risk used with systemic fibrinolysis fore, only in the presence of uPA was t-PA’scontri- are recommended. Similar results are observed when bution to fibrinolysis evident, and the importance of t-PA (0.5 to 1.0 mg/kg/h) and its mutant forms (rete- both activators in combination was also illustrated by plase and ) or tc-uPA have been used. the findings. 2102 Gurewich JACC VOL. 68, NO. 19, 2016 Improving Fibrinolysis NOVEMBER 8, 2016:2099– 106

activate it (57),followedbyreciprocalactivationof FIGURE 1 Synergistic Clot Lysis by t-PA and Pro-uPA In Vitro pro-uPA to tc-uPA, as noted (54). The high-affinity substrate binding to this plasminogen is responsible 100% t-PA + pro-uPA for pro-uPA’s fibrin specificity. Therefore, different fibrin domains promote fibrin- dependent plasminogen activation by t-PA and lysis by the combination pro-uPA. This was confirmed in a study in which Clot in 50% Human plasminogen activation was measured using the plasma isolated fibrin domain fragments. Plasminogen acti- Clot Lysis pro-uPA vationbyt-PAwasstronglypromotedonlybythe lysis by t-PA each alone fibrin fragment D-domain, whereas that by pro-uPA waspromotedonlybytheE-domain(58),reflecting B the complementary modes of action of the activators 1 2 3 4 5 h (51).Becauset-PA’s fibrinolytic function is to initiate Time fibrinolysis, this could, in time, be initiated by the intrinsic activity of pro-uPA or a little tc-uPA, as Lysis of a human plasma clot in a plasma milieu (photo) by doses of t-PA (x) or pro-uPA (o) previously shown (51), and provides an explanation in plasma that induce 25% lysis in 5 h, and clot lysis when these same doses are combined. for why a t-PA knockout had no apparent effect on In combination they induced 100% clot lysis. pro-uPA ¼ proenzyme urokinase plasmin- fibrinolysis (Central Illustration) (49,50). ogen activator; t-PA ¼ tissue-type plasminogen activator.

IN VIVO AND CLINICAL TESTING OF t-PA AND PRO-uPA IN COMBINATION Second, t-PA and pro-uPA have complementary modes of action, as each activates a different fibrin- In vitro, a synergistic effect of t-PA and pro-uPA bound plasminogen (51,52), and it is fibrin-bound combinations has long been recognized (52,53) plasmin that is responsible for fibrinolysis. The (Figure 1)andshowninvivoinanimals(59).The complementary functions oftheactivatorsalsomake combination was subsequently studied clinically in their combined effects synergistic in fibrinolysis AMIin2smallstudies.Inthefirst, 38 patients with (Figure 1) (53), and provide an explanation for the ST-segment elevation myocardial infarction were efficiency of endogenous fibrinolysis. given 12 mg of t-PA infused for 30 min and 48 mg of Third, during pro-uPA plasminogen activation, pro- pro-uPA infused for 40 min. A TIMI flow grade 3 rate uPA is reciprocally activated by plasmin to the enzyme, in the infarct artery at 90 min was seen in 57.6% of the tc-uPA (54), which promotes and completes fibrino- patients. It was concluded that the findings “support lysis. Therefore, uPA has 2 fibrinolytic functional the potential use of t-PA and pro-uPA in place of states that contribute to fibrinolysis and explain its monotherapy” (60). In the second study, 23 patients dominance because t-PA has only 1 functional state (1), with AMI were given 20 mg of t-PA and 12.3 mg of pro- consistent with the gene knockout findings (51,52).In uPA for 60 min. “Successful infarct artery patency addition, this uPA property is also in keeping with a (TIMI flow grade 2 or 3) occurred in 78% of patients” study comparing t-PA and pro-uPA clot lysis rates; that (61). It was concluded that, “combination therapy of pro-uPA was consistently twice that of t-PA (55). may allow substantial reductions in dosage” (61).In Fourth, t-PA and uPA have sequential fibrinolytic both of these studies, some bleeding complications functions. At the site of an occlusive thrombus, t-PA occurred. stored in the endothelium of the vessel wall is released, In the largest multicenter study, the PATENT (Pro- and initiates fibrinolysis by binding to its Urokinase and t-PA Enhancement of Thrombolysis) on the D-domain of intact fibrin (3) where plasminogen trial, 101 patients with ST-segment elevation is also located (4). binding of t-PA to this site myocardial infarction received a sequential combi- promotes the kinetics of its activation of this plas- nation of t-PA and pro-uPA on the basis of their minogen by about 1,000-fold (5),initiatingfibrinolysis. natural sequential functions. In the first 10 patients, a Fibrin degradation exposes new plasminogen binding 10-mg bolus of t-PA was administered to initiate sites (56),thefirst of which is an unusual triple fibrinolysis, which was followed by an infusion of carboxy-terminal lysine-binding site on the fibrin E- 40 mg/h of pro-uPA (50% of the standard AMI infu- domain. This induces a special conformational change sion rate) for 90 min. This proved to be too much t-PA in plasminogen, against which pro-uPA’s intrinsic because it caused some bleeding, as in the 2 studies activity is promoted 250-fold, enabling pro-uPA to described. Therefore, the t-PA dose was reduced to JACC VOL. 68, NO. 19, 2016 Gurewich 2103 NOVEMBER 8, 2016:2099– 106 Improving Fibrinolysis

CENTRAL ILLUSTRATION Fibrinolysis and Bleeding by t-PA and pro-uPA

Gurewich, V. J Am Coll Cardiol. 2016;68(19):2099–106.

(Top) Fibrinolysis is initiated by t-PA, which binds to its binding site on intact fibrin (D-domain) and activates the fibrin-bound plasminogen at that site to plasmin. This initiates fibrin degradation by plasmin and creates 2 additional plasminogen binding sites, the first of which, on the fibrin E-domain, induces a special plasminogen conformation that enables pro-uPA itself to activate this plasminogen to plasmin (long arrow). Plasmin induces reciprocal activation of pro-uPA to tc-uPA, which then activates the remaining plasminogen and completes fibrinolysis. (Bottom) Bleeding: hemostatic fibrin is intact fibrin and, therefore, it contains the same t-PA and plasminogen binding sites that are on the fibrin D-domain shown in the top panel. As a result, hemostatic fibrin is vulnerable to lysis by therapeutic doses of t-PA. High-dose pro-uPA activates plasma plasminogen and plasmin activates pro-uPA, generating more plasmin. pro-uPA ¼ proenzyme urokinase plasminogen activator; tc-uPA ¼ 2-chain urokinase plasminogen activator, the enzymatic form of uPA; t-PA ¼ tissue-type plasminogen activator; uPA ¼ urokinase plasminogen activator.

5 mg (5% of the standard AMI dose) for the remaining occurred in the study (62). These results compare 91 patients, with the same pro-uPA infusion, and very favorably with the best of the t-PA trials (GUSTO) there were no further bleeding episodes. in which the TIMI flow grade 3 patency at 90 min was ATIMIflow grade 3 patency of the infarct artery at 54% and decreased to 45% at 24 h, reflecting t-PA’s 90 min occurred in 60% of patients increasing to 82% reocclusion rate; there were 1.5% strokes and 6.3% at 24 h. No reocclusions or strokes and only 1 death mortality (Figure 2) (63). 2104 Gurewich JACC VOL. 68, NO. 19, 2016 Improving Fibrinolysis NOVEMBER 8, 2016:2099– 106

impression that only t-PA was involved in intravas- FIGURE 2 TIMI Flow Grade 3 Patency Rates and Mortality in the cular fibrinolysis. This concept was reinforced by the PATENT and GUSTO Trials finding that uPA binds to a cell-surface uPA receptor, TIMI-3 Patency Mortality and is responsible for extravascular plasminogen 100% 10% activation important for cell migration. It was 82% concluded that uPA was an extravascular plasminogen 75 7.5% activator (65), a concept that has tended to persist (66). 60% 6.3% Much evidence to the contrary has had little effect on 54% this concept, including the finding that a uPA cell- 50 5% 45% surface receptor knockout had no measurable effect on fibrinolysis by uPA, contrary to expectations (50). 25 2.5% Finally, because most of the uPA is located on the surface of (67,68) and monocytes (69), 1% rather than free in the plasma, it has often gone un- 90 min 24h 90 min24h 30 Day detected. All of these pro-uPA properties help to explain why its essential role in fibrinolysis has been PATENT trial GUSTO trial overlooked.

The TIMI flow grade 3 patency rates in the PATENT trial, in which a low-dose sequential CONCLUSIONS combination of t-PA (5 mg) and pro-uPA (40 mg/h) was used (62),andthoseinGUSTO,the ¼ t-PA megatrial in which the best results in AMI were obtained (15).AMI acute myocardial The 2 natural fibrin-dependent plasminogen activa- infarction; GUSTO ¼ Global Utilization of Streptokinase and Tissue Plasminogen Activator for tors have complementary fibrinolytic functions so Occluded CoronaryArteries; PATENT ¼ Pro-Urokinase and t-PA Enhancement of Thrombolysis; fi TIMI ¼ Thrombolysis In Myocardial Infarction; other abbreviations as in Figure 1. that both are needed to complete brinolysis. As a result, t-PA and pro-uPA are also synergistic in com- bination, which provides an explanation for how their MISCONCEPTIONS ABOUT uPA low physiological concentrations can induce fibrino- AND THE FIBRINOLYTIC SYSTEM lysis. In contrast, fibrinolytic therapy with high-dose t-PA has been inadequate and risky, which has Certain uPA properties and misconceptions delayed limited its use and resulted in its partial replacement its recognition and contributed to the current state of by endovascular procedures for certain indications. the art of fibrinolysis. For example, enzymatic uPA, Fibrinolysis, however, remains an essential thera- tc-uPA, was known since the 1940s as a non–fibrin- peutic option, being the only means by which rapid, specific activator, like SK, with a high bleeding risk. pharmacological removal of a thrombus is possible The proenzyme, pro-uPA, was not identified and and reperfusion can be restored at a time when the isolated until 1980 (63), and because it had no fibrin greatest benefit is achievable. A different fibrinolytic affinity, recognition of its fibrin-specific fibrinolytic regimen from t-PA monotherapy is required for this effect was delayed. Moreover, in monotherapy clin- objective, and the efficient biological paradigm of ical trials, at the concentrations required, pro-uPA in endogenous fibrinolysis provides an example that plasma converted to tc-uPA, causing bleeding com- may be difficult to improve upon. plications (12). Pro-uPA is inactivated by the clotting enzyme REPRINT REQUESTS AND CORRESPONDENCE: Dr. thrombin (64); thus, its activity was destroyed inad- Victor Gurewich, Harvard Medical School, Mount vertently in many fibrinolysis studies in which lysis AuburnHospital,300Mt.AuburnStreet,Cambridge, was preceded by thrombin clotting, giving the Massachusetts 02138. E-mail: [email protected].

REFERENCES

1. Rijken DC, Hoylaerts M, Collen D. Fibrinolytic 3. RijkenDC,CollenD.Purification and characterization 5. Hoylaerts M, Rijken DC, Lijnen HR, et al. Kinetics properties of one-chain and two-chain human of the plasminogen activator secreted by human mel- of the activation of plasminogen by human tissue extrinsic (tissue-type) plasminogen activator. anoma cells in culture. J Biol Chem 1981;56:7035–41. plasminogen activator. Role of fibrin. J Biol Chem J Biol Chem 1982;257:2920–5. 1982;257:2912–9. 4. Voskuilen M, Vermond A, Veeneman GH, et al. 2. Pannell R, Gurewich V. Activation of plasminogen by lysine residue Aa157 plays a crucial role 6. Pennica D, Holmes WE, Kohr WJ, et al. single-chain urokinase or by two-chain urokinase–a in the fibrin-induced acceleration of plasminogen Cloning and expression of human tissue-type demonstration that single-chain urokinase has a low activation, catalyzed by tissue-type plasminogen plasminogen activator cDNA in E. coli. Nature catalytic activity (pro-urokinase). Blood 1987;69:22–6. activator. J Biol Chem 1987;262:5944–6. 1983;301:214–21. JACC VOL. 68, NO. 19, 2016 Gurewich 2105 NOVEMBER 8, 2016:2099– 106 Improving Fibrinolysis

7. Rogers WJ, Baim DS, Gore JM, et al. Comparison 21. Blockmans D, Deckmyn H, Van den Hove L, response to intravenous thrombolysis for stroke. of immediate invasive, delayed invasive, and con- et al. The effect of plasmin on platelet function. Stroke 2007;38:948–54. servative strategies after tissue-type plasminogen Platelets 1996;7:139–48. 37. Emberson J, Lees KR, Lyden P, et al., Stroke activator. Results of the Thrombolysis in Myocar- 22. Winters KJ, Santoro SA, Miletich JP, et al. Thrombolysis Trialists’ Collaborative Group. Effect dial Infarction (TIMI) Phase II-A trial. Circulation Relative importance of thrombin compared with of treatment delay, age, and stroke severity on the 1990;81:1457–76. plasmin-mediated platelet activation in response effects of intravenous thrombolysis with 8. Kleindorfer D, Lindsell CJ, Brass L, et al. to plasminogen activation with streptokinase. for acute ischaemic stroke: a meta-analysis of National US estimates of recombinant tissue Circulation 1991;84:1552–60. individual patient data from randomised trials. – plasminogen activator use: ICD-9 codes substan- 23. Keeley EC, Boura JA, Grines CL. Primary Lancet 2014;384:1929 35. tially underestimates. Stroke 2008;39:924–8. angioplasty versus intravenous thrombolytic 38. IST-3 Collaborative Group. The benefits and 9. Meyer G, Vicaut E, Danays T, et al., PEITHO therapy for acute myocardial infarction: a quanti- harms of intravenous thrombolysis with recombi- Investigators. Fibrinolysis for patients with tative review of 23 randomised trials. Lancet nant tissue plasminogen activator within 6 h of – intermediate-risk pulmonary embolism. N Engl J 2003;361:13 20. acute ischaemic stroke (The Third International Med 2014;370:1402–11. 24. Stone GW, Cox D, Garcia E, et al. Normal flow Stroke Trial [IST-3]): a randomised controlled trial. – (TIMI-3) before mechanical is Lancet 2012;379:2352 63. 10. Husain SS, Lipinski B, Gurewich V. Rapid puri- an independent determinant of survival in acute fication of high affinity plasminogen activator 39. American Academy of Emergency Medicine. myocardial infarction. Circulation 2001;104: from human plasma by specific adsorption on Position Statement on the Use of Intravenous 636–41. fibrin/Celite. Proc Natl Acad Sci U S A 1981;78: Thrombolytic Therapy in the Treatment of Stroke. 4265–9. 25. DeLuca G, Ernst N, Zijlstra F, et al. Preproce- Available at: http://www.aaem.org/em-resources/ dural TIMI flow and mortality in patients with position-statements/clinical-practice/thrombolytic- 11. Pannell R, Gurewich V. Pro-urokinase: a study acute myocardial infarction treated by primary therapy. Accessed September 23, 2016. of its stability in plasma and a mechanism for its angioplasty. J Am Coll Cardiol 2004;43:1363–7. selective fibrinolytic effect. Blood 1986;67: 40. Campbell BC, Mitchell PJ, Kleinig TJ, et al., 1215–23. 26. Rapaport E. GUSTO: assessment of the pre- EXTEND-IA Investigators. Endovascular therapy liminary results. J Myocardial Ischemia 1993;5: for ischemic stroke with perfusion-imaging selec- 12. PRIMI Trial Study Group. Randomised double- 15–24. tion. N Engl J Med 2015;372:1009–18. blind trial of recombinant pro-urokinase against streptokinase in acute myocardial infarction. Lan- 27. Boersma E, Maas ACP, Deckers JW, et al. Early 41. Berkhemer OA, Fransen PS, Beumer D, et al., cet 1989;1:863–7. thrombolytic treatment in acute myocardial MR CLEAN Investigators. A randomized trial of infarction: reappraisal of the golden hour. Lancet intraarterial treatment for acute ischemic stroke. 13. Gruppo Italiano per lo Studio della Sopravvi- 1996;348:771–5. N Engl J Med 2015;372:11–20. venze nell’Infarto Miocardico. GISSI-2: a factorial 28. Schofield PM. Acute myocardial infarction: the 42. Goyal M, Demchuk AM, Menon BK, et al., randomised trial of alteplase versus streptokinase case for pre-hospital thrombolysis with or without ESCAPE Trial Investigators. Randomized assess- and heparin versus no heparin among 12,490 pa- percutaneous coronary intervention. Heart 2005; ment of rapid endovascular treatment of ischemic tients with acute myocardial infarction. Lancet 91 Suppl 3:iii7–11. stroke. N Engl J Med 2015;372:1019–30. 1990;336:65–71. 29. Reimer KA, Vander Heide RS, Richard VJ. 43. Kearon C, Akl EA, Comerota AJ, et al. Antith- 14. ISIS-3 (Third International Study of Infarct Reperfusion in acute myocardial infarction: effect rombotic therapy for VTE disease: Survival) Collaborative Group. ISIS-3: a rando- of timing and modulating factors in experimental Therapy and Prevention of Thrombosis, 9th ed: mised comparison of streptokinase vs tissue models. Am J Cardiol 1993;72:13G–21G. American College of Chest Physicians Evidence- plasminogen activator vs and of Based Clinical Practice Guidelines [Published plus heparin vs aspirin alone among 41,299 30. Keeley EC, Boura JA, Grines CL. Comparison of correction appears in: Chest 2012;142;1698–704]. cases of suspected acute myocardial infarction. primary and facilitated percutaneous coronary Chest 2012;141:e419S–494. Lancet 1992;339:65–71. interventions for ST-elevation myocardial infarc- tion: quantitative review of randomised trials. 44. Enden T, Haig Y, Kløw NE, et al., CaVenT 15. GUSTO Angiographic Investigators. The effects Lancet 2006;367:579–88. Study Group. Long-term outcome after additional of tissue plasminogen activator, streptokinase, or 31. Weaver WD, Hartmann JR, Anderson JL, et al. catheter-directed thrombolysis versus standard both on coronary artery patency, ventricular New recombinant glycosylated prourokinase for treatment for acute iliofemoal deep vein throm- function, and survival after acute myocardial bosis (the CaVenT Study): a randomised controlled – treatment of patients with acute myocardial infarction. N Engl J Med 1993;329:1615 22. – infarction. J Am Coll Cardiol 1994;24:1242–8. trial. Lancet 2012;379:31 8. 16. Brophy JM, Joseph L. Placing trials in context 32. Stone GW, Gersh BJ. Facilitated angioplasty: 45. Morrison HL III. Catheter-directed thrombol- using Bayesian analysis. GUSTO revisited by Rev- paradise lost. Lancet 2006;367:543–6. ysis for acute limb ischemia. Semin Intervent – erend Bayes. JAMA 1995;273:871 5. Radiol 2006;23:258–69. 33. Assessment of the Safety and Efficacy of a 17. Verheugt FWA, Meijer A, Lagrand WK, et al. New Treatment Strategy with Percutaneous Cor- 46. Plate G, Oredsson S, Lanke J. When is fl Reocclusion: the ip side of coronary thrombol- onary Intervention (ASSENT-4 PCI) investigators. thrombolysis for acute lower limb ischemia – ysis. J Am Coll Cardiol 1996;27:766 73. Primary versus tenecteplase-facilitated percuta- worthwhile? Eur J Vasc Endovasc Surg 2009;37: 206–12. 18. Rapold HJ, Grimaudo V, Declerck PJ, et al. neous coronary intervention in patients with ST- Plasma levels of plasminogen activator inhibitor segment elevation acute myocardial infarction 47. Chandler WL, Jascur ML, Henderson PJ. type 1, b-thromboglobulin, and fibrinopeptide A (ASSENT-4 PCI): a randomised trial. Lancet 2006; Measurement of different forms of tissue plas- – before, during and after treatment of acute 367:569 78. minogen activator in plasma. Clin Chem 2000; myocardial infarction with alteplase. Blood 1991; 34. Boysen G, ECASS Study Group. European 46:38–46. 78:1490–5. Cooperative Acute Stroke Study (ECASS): (rt-PA- 48. Baglin TP, Langdown J, Frasson R, et al. 19. Eisenberg PR, Sherman L, Rich M, et al. Thrombolysis in acute stroke) study design and Discovery and characterization of an antibody – Importance of continued activation of thrombin progress report. Eur J Neurol 1995;1:213 9. directed against exosite I of thrombin. J Thromb reflected by fibrinopeptide A to the efficacy of 35. Rha JH, Saver JL. The impact of recanalization Haemost 2016;14:137–42. thrombolysis. J Am Coll Cardiol 1986;7:1255–62. on ischemic stroke outcome. A meta-analysis. 49. Bugge TH, Flick MJ, Danton MJ, et al. Stroke 2007;38:967–73. 20. Owen J, Friedman KD, Grossman BA, et al. Urokinase-type plasminogen activator is effective Thrombolytic therapy with tissue plasminogen 36. Saqqur M, Uchino K, Demchuk AM, et al., in fibrin clearance in the absence of its receptor or activator or streptokinase induces transient CLOTBUST Investigators. Site of arterial occlusion tissue-type plasminogen activator. Proc Natl Acad thrombin activity. Blood 1988;72:616–20. identified by transcranial Doppler predicts the Sci U S A 1996;93:5899–904. 2106 Gurewich JACC VOL. 68, NO. 19, 2016 Improving Fibrinolysis NOVEMBER 8, 2016:2099– 106

50. Singh I, Burnand KG, Collins M, et al. Failure of degraded fibrin I. J Biol Chem 1985;260: 63. Husain SS, Gurewich V, Lipinski B. Purification thrombus to resolve in urokinase-type plasmin- 4432–40. and partial characterization of a single-chain, ogen activator gene-knockout mice: rescue by high molecular weight form of urokinase from 57. Liu J, Gurewich V. Fragment E-2 from fibrin normal bone marrow-derived cells. Circulation human urine. Arch Biochem Biophys 1983;220: substantially enhances pro-urokinase-induced 2003;107:869–75. 31–8. Glu-plasminogen activation. A kinetic study using 51. Gurewich V. Experiences with pro-urokinase a plasmin-resistant mutant pro-urokinase 64. Gurewich V, Pannell R. Inactivation of single and potentiation of its fibrinolytic effect by uro- (Ala-158-rpro-UK). Biochemistry 1992;31:6311–7. chain urokinase (pro-urokinase) by thrombin and and by tissue plasminogen activator. J Am thrombin-like : relevance of the finding to 58. Liu J, Gurewich V. A comparative study of the Coll Cardiol 1987;10:16B–21B. the interpretation of fibrin binding experiments. promotion of tissue plasminogen activator and Blood 1987;69:769–72. 52. Pannell R, Black J, Gurewich V. The comple- pro-urokinase-induced plasminogen activation by mentary modes of action of tissue plasminogen fragments D and E-2 of fibrin. J Clin Invest 1991; 65. Longstaff C, Kolev K. Basic mechanisms and activator (t-PA) and pro-urokinase (pro-UK) by 88:2012–7. regulation of fibrinolysis. J Thromb Haemost 2015; which their synergistic effect on clot lysis may be 13 Suppl 1:S98–105. 59. Collen D, Stassen JM, Stump DC, et al. explained. J Clin Invest 1988;81:853–9. Synergism of thrombolytic agents in vivo. Circu- 66. Collen D. The plasminogen (fibrinolytic) 53. Gurewich V, Pannell R. Synergism of tissue- lation 1986;74:838–42. system. Thromb Haemost 1999;82:259–70. type plasminogen activator (t-PA) and single- 60. Bode C, Schuler G, Nordt T, et al. Intravenous 67. Vaughan DE, Van Houtte E, Collen D. Urokinase chain urokinase-type plasminogen activator thrombolytic therapy with a combination of binds to platelets through a specific saturable, low (scu-PA) on clot lysis in vitro and a mechanism for single-chain urokinase-type plasminogen activator affinity mechanism. Fibrinolysis 1990;4:141–6. this effect. Thromb Haemost 1987;57:372–3. and recombinant tissue-type plasminogen acti- 68. Gurewich V, Johnstone M, Loza JP, et al. Pro- 54. Petersen LC. Kinetics of reciprocal pro- vator in acute myocardial infarction. Circulation urokinase and are both associated urokinase/plasminogen activation–stimulation by 1990;81:907–13. with platelets. Implications for the intrinsic a template formed by the 61. Kirshenbaum JM, Bahr RD, Flaherty JT, et al. pathway of fibrinolysis and for therapeutic bound to poly(D-lysine). Eur J Biochem 1997;245: Clot-selective coronary thrombolysis with low- thrombolysis. FEBS Lett 1993;318:317–21. 316–23. dose synergistic combinations of single-chain 69. Grau E, Moroz LA. Fibrinolytic activity of 55. Pannell R, Gurewich V. A comparison of the urokinase-type plasminogen activator and normal human blood monocytes. Thromb Res rates of clot lysis in a plasma milieu induced by recombinant tissue-type plasminogen activator. 1989;53:145–62. tissue plasminogen activator (t-PA) and rec-pro- Am J Cardiol 1991;68:1564–9. urokinase: evidence that t-PA has a more 62. Zarich SW, Kowalchuk GJ, Weaver WD, et al. restricted mode of action. Fibrinolysis 1992;6:1–5. Sequential combination of thrombolytic therapy KEY WORDS endogenous lysis, 56. Harpel PC, Chang TS, Verderber E. Tissue for acute myocardial infarction: results of the plasminogen activator inhibitor, plasminogen activator and urokinase mediate the Pro-Urokinase and t-PA Enhancement of streptokinase, synergy, tissue plasminogen binding of Glu-plasminogen to plasma fibrin I. Thrombolysis (PATENT) trial. J Am Coll Cardiol activator, urokinase-type plasminogen Evidence for new binding sites in plasmin- 1995;26:374–9. activator